Quantitative extraction of high-dimensional mineable data from medical images is a process known as radiomics. Radiomics is foreseen as an essential prognostic tool for cancer risk assessment and the quantification of intratumoural heterogeneity. In this work, 1615 radiomic features (quantifying tumour image intensity, shape, texture) extracted from pre-treatment FDG-PET and CT images of 300 patients from four different cohorts were analyzed for the risk assessment of locoregional recurrences (LR) and distant metastases (DM) in head-and-neck cancer. Prediction models combining radiomic and clinical variables were constructed via random forests and imbalance-adjustment strategies using two of the four cohorts. Independent validation of the prediction and prognostic performance of the models was carried out on the other two cohorts (LR: AUC = 0.69 and CI = 0.67; DM: AUC = 0.86 and CI = 0.88). Furthermore, the results obtained via Kaplan-Meier analysis demonstrated the potential of radiomics for assessing the risk of specific tumour outcomes using multiple stratification groups. This could have important clinical impact, notably by allowing for a better personalization of chemo-radiation treatments for head-and-neck cancer patients from different risk groups.

No abstract

No abstract

IntroductionMatrix metalloproteinase (MMP)-2 is very active at degrading extracellular matrix. It is under the influence of an activator, membrane type 1 MMP (MMP-14), and the tissue inhibitor of metalloproteases (TIMP)-2. We hypothesized that the individual expression of these three markers or their balance may help to predict breast cancer prognosis.MethodsMMP-2, MMP-14 and TIMP-2 expression has been evaluated by 35S mRNA in situ hybridization on paraffin material of 539 breast cancers without distant metastasis at diagnosis and with a median follow-up of 9.2 years.ResultsMMP-2 and MMP-14 mRNA was detected primarily in reactive stromal cells whereas TIMP-2 mRNA was expressed by both stromal and cancer cells. Of the three molecules, an adjusted Cox model revealed that high MMP-14 mRNA (≥ 10% cells) alone predicted a significantly shorter overall survival (p = 0.031) when adjusted for clinical factors (tumor size and number of involved lymph nodes). Prognostic significance was lost when further adjusted for Her-2/neu and urokinase-type plasminogen activator (p = 0.284). Furthermore, when all three components were analyzed together, the survival was worst for patients with high MMP-2/high MMP-14/low TIMP-2 (5 year survival = 60%) and best with low MMP-2/low MMP-14/high TIMP-2 (5 year survival = 74%), but the difference did not reach statistical significance (p = 0.3285).ConclusionOf the MMP-14/TIMP-2/MMP-2 complex, MMP-14 was the factor most significantly associated with the outcome of breast cancer and was an independent factor of poor overall survival when adjusted for clinical prognostic factors, but not for certain ancillary markers.

The analysis of study 1 shows a benefit of using a short course of neoadjuvant AS with EBRT vs EBRT alone for localized T2-T3 prostate cancers. Moreover, in each study adding a short course of adjuvant AS after neoadjuvant 1 provided no more advantage in these patients.